Improved manifold coordinate representations of large-scale hyperspectral scenes

被引:122
作者
Bachmann, Charles M. [1 ]
Ainsworth, Thomas L. [1 ]
Fusina, Robert A. [1 ]
机构
[1] USN, Res Lab, Remote Sensing Div, Washington, DC 20375 USA
来源
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 2006年 / 44卷 / 10期
关键词
automatic classification; hyperspectral imagery; isometric mapping (ISOMAP); Jeffries-Matsushita distance; manifold coordinates; manifold geodesics; manifold learning; multidimensional scaling; nonlinear dimensionality reduction; tree searching; trees (graphs); Vantage Point Forest; vantage point tree; Virginia Coast Reserve;
D O I
10.1109/TGRS.2006.881801
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
In recent publications, we have presented a data-driven approach to representing the nonlinear structure of hyperspectral imagery using manifold coordinates. The approach relies on graph methods to derive geodesic distances on the high-dimensional hyperspectral data manifold. From these distances, a set of intrinsic manifold coordinates that parameterizes the data manifold is derived. Scaling the solution relied on divide-conquer-and-merge strategies for the manifold coordinates because of the computational and memory scaling of the geodesic coordinate calculations. In this paper, we improve the scaling performance of isometric mapping (ISOMAP) and achieve full-scene global manifold coordinates while removing artifacts generated by the original methods. The CPU time of the enhanced ISOMAP approach scales as O(N log(2) (N)), where N is the number of samples, while the memory requirement is bounded by O(N log (N)). Full hyperspectral scenes of O(10(6)) samples or greater are obtained via a reconstruction algorithm, which allows insertion of large numbers of samples into a representative "backbone" manifold obtained for a smaller but representative set of O(10(5)) samples. We provide a classification example using a coastal hyperspectral scene to illustrate the approach.
引用
收藏
页码:2786 / 2803
页数:18
相关论文
共 52 条
[21]  
BOZKAYA T, 1997, P ACM SIGMOD INT C M, P357
[22]   Pivot selection techniques for proximity searching in metric spaces [J].
Bustos, B ;
Navarro, G ;
Chávez, E .
PATTERN RECOGNITION LETTERS, 2003, 24 (14) :2357-2366
[23]  
Chen YC, 2005, INT GEOSCI REMOTE SE, P4311
[24]  
Cormen T. H., 2001, Introduction to Algorithms, V2nd
[25]   Ocean PHILLS hyperspectral imager: design, characterization, and calibration [J].
Davis, CO ;
Bowles, J ;
Leathers, RA ;
Korwan, D ;
Downes, TV ;
Snyder, WA ;
Rhea, WJ ;
Chen, W ;
Fisher, J ;
Bissett, WP ;
Reisse, RA .
OPTICS EXPRESS, 2002, 10 (04) :210-221
[26]  
Davis S. M, 1978, Remote Sensing: The Quantitative Approach
[27]  
Dijkstra E. W., 1959, NUMER MATH, V1, P269, DOI DOI 10.1007/BF01386390
[28]  
Duda R.O., 2001, Pattern Classification, V2nd
[29]  
Dudani S. A., 1976, IEEE Transactions on Systems, Man and Cybernetics, VSMC-6, P325, DOI 10.1109/TSMC.1976.5408784
[30]  
*ENVI IDL, ITT VIS INF SOL